Prenatal exposure to Per- and polyfluoroalkyl substances and adiposity measures of children at 4 and 6 years: A prospective birth cohort in China

There is growing evidence that prenatal exposure to Per-and polyfluoroalkyl substances (PFAS) was associated with childhood obesity, but evidence on multiple adiposity measures including arm circumference (AC), and waist circumference (WC) among Chinese children is limited. We investigated the associations of prenatal exposure to PFAS with adiposity measures of children at 4 and 6 years of age in the Shanghai-Minhang Birth Cohort Study. A total of 573 mother-child pairs with maternal PFAS concentrations and at least one measurement of adiposity measures of children were included in the present study. Eleven PFAS were assessed in maternal fasting blood samples. Information on children's weight, height, AC, and WC was collected at follow-ups. Weight for age Z score (WAZ), body mass index for age Z score (BMIz), and children overweight were calculated based on the World Health Organization Child Growth Standards. Multivariate linear regression, Poisson regression with robust error variance, and Bayesian Kernel Machine Regression (BKMR) models were used to examine the as-sociations of prenatal exposure to PFAS with children's adiposity measures. Eight PFAS with detection rates above 85 % were included in the analyses. In the multivariate linear regression models, maternal PFNA con-centrations were associated with a greater AC (8 = 0.29, 95 % Confidence Interval (CI): 0.04-0.55) in 4-year-old children and with an increase in WAZ (8 = 0.26, 95 % CI: 0.06-0.46), BMIz (8 = 0.31, 95 % CI: 0.09-0.53), AC (8 = 0.49, 95 % CI: 0.08-0.90), and WC (8 = 1.47, 95 % CI: 0.41-2.52) in 6-year-old children. We also observed the associations of maternal concentrations of PFOS, PFNA, PFUdA, and PFTrDA with the increased risk of children overweight in 6-year-old children. BKMR models further supported the findings from multivariate linear regression and Poisson regression models, and identified PFNA as the most important contributor. Moreover, the associations described above were generally more pronounced in girls. In conclusion, prenatal exposure to PFAS was associated with an increased risk of children's adiposity with a sex-specific manner, and PFNA contributed most to the associations after controlling for the effect of co-exposure to other PFAS compounds, especially among girls at 6 years of age.